Semantic analysis of a sentence. Reflections about the meaning of the meaning and the Artificial Intelligence

Analysis of the meaning of a sentence, based on the knowledge base of an operational thinking machine. Reflections about the meaning and artificial intelligence

By Todor Arnaudov | 13 March 2004 @ 21:49 EET | 340 reads | First published at bgit.net and the e-zine “Sacred Computer”

Оригиналът на български: http://artificial-mind.blogspot.com/2008/02/2004.html или: Анализ на смисъла на изречение въз основа на базата знания на действаща мислеща машина. Мисли за смисъла и изкуствената мисъл. Една от основополагащите ми статии от тийнейджърските ми години. Тази е от късните, продължава с два дълги коментари, добавки, включващи и размисли върху принципи на ученето с подкрепление (reinforcement learning)

Included in “Unverse and Mind 4": a search of the meaning (http://eim.hit.bg/razum, in Bulgarian only yet) Finally I started to translate my old articles, including my teenage Theory of Mind and Universe - all were milestones of my AGI research. I wrote this particular one as a 19-year old freshman in Computer Science at Plovdiv University.

• Natural Language Processing, NLP.
• An example of a search through the linguistic knowledge base of the author.
• Many different meanings of "meaning" defined for different uses.
• Quasi-formal semantic analysis - from words to chunks (expressions) to clauses to complex sentences. A search for relations, links.
• Criticism of the usage of short ambiguous sentences as a way to explain the "impossibility" of creation of an AI and machine translation; the lacking context is filled by human imagination and could be filled with machine's imagination. A discussion about the artificial pruning of the set of interpretations that humans are doing when translating or interpreting, and the implicit denial of expecting unknown meanings.
• Thought experiment of how a 3-year old toddler boy interprets unknown sentence (tme flies) and how does he search for a meaning and maps meanings to his senses. It is told as a real story, what he knows, what he experiences and what he would experience if...
• The style in not really academic, there are some dialogues, discussions with a virtual opponent, when this is appropriate to display human biases.
• Others...
• Continues with long comments with additions about reinforcement learning and other topics (to be translated and linked)

Part 1 (и български):(This Post) Semantic analysis of a sentence. Reflections about the meaning of the meaning and the Artificial Intelligence 

Part 2 (и български): Causes and reasons for human actions. Searching for causes. Whether higher or lower levels control. Control Units. Reinforcement learning. 

Part 3 (и български): Motivation is dependent on local and specific stimuli, not general ones. Pleasure and displeasure as goal-state indicators. Reinforcement learning.

Part 4 : Intelligence: search for the biggest cumulative reward for a given period ahead, based on given model of the rewards. Reinforcement learning.

Some of the conclusions:

• More complex (smarter) the system – More Meaningless the Meaning

Because the interpretation depends more on the system.

• „What is possible to be done with” is yet another meaning of the concept of “meaning”.

Happy reading and enjoy the story of the little Johny and the flies that are flying around the watch... ;)

BEGIN...

This article starts with the definition of “meaning” of a friend of mine, Ilian Georgiev, we shared thoughts an year ago.

Meaning/Sense (Ilian Georgiev's definition): Meaning of a sentence (a thought) is a function that searches for controversy between the knowledge base of the evaluator (the one who thinks) and the thought being analyzed. All elements from the knowledge base and their connections are juxtaposed with the new thought, the one that is analyzed. If any of the elements of the thought has a connection with another element, and the connection cannot be found in the knowledge base, then the sentence is classified as a non-sense (has an error).

Using this definition, let's search for the meaning of a weird sentence, that I made up don't know how. I'll use also a made-up semi-formal syntax of NLP analysis, that you'll grasp on-the-fly.

(The original paper is in Bulgarian, and the author is not a native English speaker; it is possible that there are some mistakes in some of the senses used.) SENTENCE = “The cat drank the stone and flew out under the uphill.” ИЗРЕЧЕНИЕ == "Котката изпи камъка и литна под нанагорнището."

SENTENCE == SENTENCE_1 + SENTECNE_2 (clauses)

SENTENCE_1 == "The cat drank the stone" 

- Agent_1 == The cat

- Action == drank 

- Object == the stone

SENTENCE_2 == "Литна под нанагорнището". 

- Agent + Preposition + Object --

- Agent_2 == Agent_1 - Preposition == under - Object == the uphill Basic syntactic parsing check passes. Moving to the Semantic Analysis. 

  Semantic Analysis

This is a quasi-formal sample semantic analysis, using author's personal linguistic knowledge base An improvised approach 1. Word-by-word semantic analysis "Котката изпи камъка и литна под нанагорнището."

The cat drank the stone and flew out under the uphill.

Do I know what does mean:
 -- the cat -- :: YES, a defined “cat”, I know what “cat” means
 -- drank -- :: YES, pass tense of the verb “to drink”
 -- the stone -- :: YES, Noun, ...
 -- и -- :: YES, Conjuction
 -- flew out -- :: YES, Verb, past tense
 -- under -- :: YES, Preposition
 -- the uphill -- :: YES, Noun

This stage passes, connections are found in the knowledge base.

2. Chunks, Expressions, Multi-word Semantic Analysis

2.1. Two words, except cases where the first word is Noun and the second is a Conjunction.

– the cat drank ...- YES (AGENT + VERB)
-- (Someone) drank the stone - YES (AGENT + VERB + NOUN/OBJECT)
-- and flew out -- YES (CONJUNCTION + VERB)
-- flew out under -- YES (VERB + PREPOSITION)
-- under the uphill -- YES (PREPOSITION + NOUN/OBJECT)

Passed.

3. Semantic analysis of the clauses and the whole sentence. "Котката изпи камъка и литна под нанагорнището." Let's analyze all the clauses, this would give clues for the meaning of earlier clauses and the whole sentence.

„The cat drank the stone” Is "the cat" linked to the verb “to drink”? - YES. What are the links? - Usually “to drink” is linked with objects, which are linked to liquidity or semi-liquidity of a substance. In general, in the definition of an object, that is linked to the verb “to drink”, usually there's a morpheme or semantics of liquidity.

“Liquid” is linked to:

1. Flow
2. Run
3. Stream down
4. Roll down
5. Trickle down
6. Pour out
7. Pour into
8. Infuse

... „To Drink” is linked specifically with:
1. To drink + water.
2. To drink + juice.
3. To drink + tea.
4. To drink + cold tea.
5. To drink + alcohol.
6. To drink + beer.
7. To drink + wine.
8. To drink + scotch.
9. To drink + vodka.
10. To drink soup. ...

The examples are checked easier if they are put down in a unified way: Verb + Object/Noun for quick comparisons. I don't know whether to check the general concepts/meanings first (like liquidity), or after all specific cases are checked first (as colocations, like “drink vodka”). „Drink” is used in some other cases, where the object is not a liquid.

1. Drink some poison.
2. Take a pill. (Take == Drink)
3. Take a medicine. (Take == Drink)

In this cases the object is:
1. On the surface of a liquid.
2. Floats in a liquid.
3. Dropped in a liquid.
4. Sunk in a liquid.
5. Absorbed in a liquid. The most common meaning of “to drink” is linked with an AGENT which is a living being. Living beings have a throat, where the drunk object passes. The liquid assists the object to pass through the throat, when the object is not a liquid itself. I recall an idiom (Bulgarian, this is a literal translation) "A duck has drunk his sense."

So, what about the linkage between “to drink” and the object “stone”? Stone has direct links to verbs as "to throw” and “to crack” and similar, “to kick” and others. Direct links are examples of usage which I have ever encountered in texts or speeches. „Stone” can play different roles: AGENT (subject) or an object, that clarifies/specifies an ACTION, done by another AGENT. Is it possible to drink a stone?

One can assume also, that if something is said, then this is a special kind of stone. If a meaning should be found in any price, it is possible for the searcher to invent meaning that matches the given sentence. It is possible also to add additional sense, using experience, so that the sentence that is “meaningless” up to now to get its explanation. This point will be discussed again later.

STONE, without a preposition

1. Kick a stone
2. Throw a stone
3. Push a stone.
4.... Lift, Roll, Hit, Crack, Break, Catch, Leave, Heat, ….

Etc... but “To drink a stone” is lacking.

However, this doesn't make the sentence meaningless, yet, because it was found above, that “drink” could be linked to objects which are not liquid, and in this sense, “to drink something” is a reference to “to swallow something” (to pass it through the throat) Drink == Swallow

Therefore: “The cat drank the stone” == “The cat swallowed the stone”

Therefore this clause makes sense according to my KB. Next clause: "...flew out under the uphill" Cat + fly?... No basic connections... Cat + jump, scratch, bite, drink, eat, … push, walk, run, fall, hide, pull, climb, jump over, hit, meaow, stalk, ... „Jump”, “Jump over” and “fall” captured my attention:

"Jump” is an action where the AGENT reaches to a state, where its body doesn't touch the ground.

It is the same for “Fly”. Therefore, “to jump” partially covers the meaning of “to fly”. Besides, I know examples of “flying” where to fly is used with the sense of “to jump” - directly or implicitly suggested by the context.

“Air Jordan” (in Bulgarian - "Въздушният Майкъл Джордан.") The basketball player Michael Jordan jumps and stays in the air for long enough to impress people more than the typical jumpers, this has caused his jumps to be linked with the morpheme “air”, which is used in words for flying (airplane, air force).

Therefore “(the cat) flew below the uphill” can be interpreted as:

(the cat) jumped below the uphill. Now, is it possible “jump” to be used with “below”. Can you jump below? To.. - Jump over s.t. - Jump into - Jump out

...

No “jump below”, but it doesn't mean that this expression is meaningless. (Actually there is jump below something, but say, not jump below the uphill) „Jump” has other meanings, like: doing something faster than usual, or moving fast. So... “[The cat] flew below the uphill” may mean:

[The cat] jumped below the uphill? Is it possible to jump below an uphill? Why not?

Uphill is an object, it can be located in а mountain, but we can imagine it to be any other object, over which somebody can move “up”, walking on. This “uphill” object can be made of wood or metal and can have a hollow inside, where a cat can hide.

The cat jumped under the thing, that had an uphill over itself... The whole sentence can turn to: "Котката изпи камъка и литна под нанагорнището." "The uphill had a hollow inside. The cat swallowed the stone and jumped below it. The additional clause fills up the uncertainty in the scenario.

Another interpretation, based on “to fall” could turn: "...flew below the uphill" to “..felt below the uphill". Also, “below the uphill” may mean below the part of the uphill, that is steep, i.e. just before the uphill starts to climb. Then: "The cat swallowed the stone and felt below the uphill." This gives a rise of another interpretation – stone is often linked with “heavy”. There is a proverb (Bulgarian) “Hang a stone on my neck”. Then: "The cat was climbing the uphill, but it swallowed the heavy stone - it threw it down below the uphill...” Or: While the cat was climbing the uphill, weird little balls felt down from the sky. They seemed like meet balls and smelled the same way. The poor cat was tired of hunger and the hard walk, and she bit one of the sky meet balls. A moment after she was frightened – the meet ball appeared to be as heavy as a stone. The cat was rolling down, until she stopped on the flat land under the uphill.

When searching for a meaning in very short pieces of information, such as single sentences, it is expected for the mind to invent, to imagine in order to fill up what is unknown with probable sets of circumstances. If the source doesn't deny, we can invent any plausible imagined circumstances. Short sentences as a way to deny creation of AI Actually, one very rarely meets single sentences in the reality, out of context to constraint and direct translation or interpretation of the meaning. However, short “nonsenses” – which one cannot interpret unambiguously or are often used to disprove the possibility of creation of an AI. Let's check out a classic from the NLP. Time flies. "It's so hard to translate to another language!”

How would we translate it in Bulgarian? "Времето лети" (Vremeto leti - The time is flying) or „Времеви мухи" (Vremevi muhi - Flies which are related to time)

Or another way - "time" doesn't mean only “time”, and “flies” doesn't mean only the little flying bug. These are just the first two items that came up to my mind! The search was obviously had been pruned up to two items, two possible interpretations.

This kind of unconscious pruning, limitation of the number of variants, will be discussed below.

Virtual Colleague: “The time is flying” is the correct translation. There's neither such an expression as “flies, related to time”, nor any other.

Author: Why do you think so?

V. Colleague: The other translations don't make any sense. Me myself, I would translate it that way, I think I'm good enough in English. Check out my personal web page. 

 Author: And why would you translate it that way? 

 Colleague: Because... I haven't heard of an expression meaning “flies, related to time”... 

  Author: Therefore you have excluded the possibility to hear a new sentence, where this expression is used in a meaning that was unknown for you before? 

  Colleague: Well, I think so... 

  Author: Who did tell you that sentence? 

  Colleague: I'm not quite sure... You? But... Well... It is possible that this is an idiom. Can you explain it to me? Maybe this is a special kind of flies? Or more likely... (What a SF fan I am not to guess this one!) Flies through time! Time travel! The ambiguity is caused by the lack of a criterion for pruning. Until the moment when an action is executed – an action caused by the input data, which are said to be ambiguous – the ambiguity is not an issue. The system can remember the whole sentence, word-by-word and until the moment of action, a decisive single action, it is known that all interpretations are possible.

And when the action should be done, e.g. a robot to capture the right cube or the middle cylinder – then the system should use an additional feature in order to disambiguate, to choose. However, since the input is not decisive, but ambiguous, then turning any of the interpretations to action is not a “mistake”, regarding the input.

Perhaps the Natural Language, or as the author calls it – The Language of Mind – allows ambiguity, because there are many “correct” possibilities. There are many cases, where each of the possible solutions/interpretations is “right” in the sense that the device (human) who took the decision continues to function after executing in effect, for real, an action, caused by the given interpretation.

So, if a given system continues to function – according to a given definition of “functions”, e.g. its heart continues to work at least for so-and-so long period after executing the given action – then this action was “right”, i.e. this action is assumed to have had followed laws that don't lead to malfunction. More complex control units have larger space of correct decisions, they have wider “freedom”, i.e. possibilities for future actions, after which they will continue to function right.

(see … @ quote “Conception for the Universal Predetermination”, a.k.a. “The Universe-Computer” or “The Mind and Universe”)

По-сложните управляващи устройства (виж "Схващане за всеобщата предопределеност": http://eim.hit.bg/razum).

I think that everything makes sense, colleagues. One can always find meaning/sense, i.e. a connection between items. The meaning is the connection between things. (The relations between things)

The easiest thing to do is to redraw an already known, drawn line, and this is what is done initially when one is doing a semantic check – whether precomputed links/connections/relations with the given expression do exist. If such links do exist, they are seen like “gray lines”, which mind can darken, one can do it when given a piece of paper with gray lines and is being told to draw lines without thinking a lot or planning. This is what the virtual colleague did above, he rejected the possibility that “time flies” has meanings, that are yet unknown to him, and need to be computed, “drawn” in his memory. Let's overview a case with the same sample expression in another case.

Time Flies... A three year-old little native English speaker – Johny. He knows, that “a fly” means the flying bug (something little, black, that is flying and when it land on your face it's !!! гъделичка and you're trying to let it go by waving your hands.

Johny knows how to create a multiple of fly – flies, but he doesn't know that “a fly” means also “a fly of an airplane”. For Johny, “time” means just “a watch”. Johny knows, that “a clock” and “a watch” have similar meanings – something circular, with a long things, that are rotating... and the longer things are rotating faster than the shorter and the thicker; the thicker ones sometimes appear not to move at all, but after you have played for a while with your toy cars and look to them – they seemed to be at another place...

All the times when Johny has heard talks about time, he has seen clocks or watches.

Johny has heard his father saying “I don't have time, we have to hurry up!” and when his father has told that, he has looked to his watch.

That way, the conception of “time” is linked to the image of “watch”, when hearing time, he sees a watch, no abstract concepts. Johny himself doesn't have a watch.

Now let's assume that we put on our hand a big and shiny colourful watch and go to play with Johny on the playground. What he is going to do, if we tell him “Time flies!” and he hears this for the first time in his life?

Colleague: Perhaps he will look to our shiny watch and will search for flies around it... 

Author: Exactly! Can you imagine what he would do if we didn't have a watch on our wrist? Colleague: Maybe he would look to our hand, searching for a watch and flies... If he has remembered the pattern of watches being on the left wrist, he may first check there, or he may check both... 

The images Johny has for “time” and “flies” are recalled, and Johny searches the expression of these images in the environment, accessible by his senses. The specific mean is not important.

The machine needs an external environment, where to search for meaning and senses – MATCHES of images, names, features, coincidences, patterns.

Author: What Johny is going to do, after realizing that there are no flies around the watch, or even there is not a watch? Colleague: It depends what behavioral models have been developed so far. He could remember the expression "time flies" as an image that represent what he has thought then - “flies which are flying around a watch”, but not to do anything further; Johny could wait, expect to face usage of this expression in an environment, which is richer of details and features, so that he would be able to extract or approve the meaning.

Details, specific cases are what limits the space of search, the domain. Details are forces for pruning...

Author: Johny may also not make any conclusion, but just taking the expression as a non-sense so he wouldn't remember it. Also, he can ask us immediately: - What does “time flies” mean? The machine should also be able to do like that, as we do, it will need teachers and supervisors, while it develops.

Our explanations and the degree of trust he has to what we explain to him will determine how the child is going to limit the space of search, but also how he will expand the space, by adding possibilities which he didn't thought of before.

If one explains to Johny, that “time flies” means “time is never enough”, the child may remember this explanation as a whole sentence, without interpretation. Just a reference, a link: “time flies” redirects to “time is never enough” and then he would search for a meaning for the new sentence. On the other hand, one can also explain to Johny, that “flies” means also “to fly”, to move like birds or like Superman or so (recall that he didn't know the verb; it's strange not to know it, but that's the assumption), but not explaining him about the abstract concept of time.

In this case, Johny could keep linking “time” to “watch” and may start to imagine “time flies” as “the watch flies”. He may look around, searching for a watch that is flying – generally, this is a search of features, input data/senses which could confirm the link that was made. Or... just anytime he hears “time flies” Johny would imagine a flying watch and would ask himself “Whether the flying watches have wings or they are magically flying?”

Imagination is a the point here. When searching for a meaning, we should be able to imagine, to fantasize. That means, one kind of inputs/senses to cause other kinds of inputs/senses. The primary may be “real”, taken from raw data from the reality, linked to what the machine or human takes for “Reality”; while the secondary input/sense could be imagined, fantasized, unreal. Talking about reality, humans usually take for “real” input channels such as vision, hearing, touch, taste, smell; when he is receiving data at the maximum possible rate (max resolution, raw data). Vision is a primary sensory input, when we're sensing images, where we can recognize individual pixels. Letters, numbers and any symbols come from a secondary sensory input, because from the primary sensory input, containing raw pixels, each one containing an independent value, are extracted data with a smaller size (in raw bits) – letters, digits, geometric shapes etc. ... In order a system to find a meaning and make sense of things, it is very useful the system to have at least two different kinds of sensory inputs, and each of them to be able to invoke, to link to the other one. The images and relations between images (e.g. motion); sounds and relations between them can “generate” words: interpretations, which are described with smaller quantity of bits. When Johny hears “a watch”, he can imagine, somehow to see the image of the watch and what is possible to be done with it.

„What is possible to be done with” is yet another meaning of the concept of “meaning”. There is no sense in meaning, if you can't do anything with it.

Actually, everything “makes sense”, or “has a meaning”, in the sense that it causes something to be done. It is so, because even the so called “non-procedural knowledge/data” are only “non-procedural”, non-active, in the sense that they are not causing a type of action that is formally defined as “procedural”. In a computer, an information processing system, any data cause actions. Data defines what happens in the machine's “mind”. For example, the descriptive data of a page of text determines what exactly the machine is supposed to do, while it's CPU is reading the data from the memory, processes it, displays it, prints the page.

The meaning is the action, that the “thing” which is evaluated invokes/causes/turns, and this meaning can be different, depending how deep we are searching and more – how and what.

How and What we do the search is more dependent on our past experience, than on the short input being evaluated. Finally... Let's assume that the meaning of a message/sentence is an action to be taken, assuming that it was caused by the meaning that was found (if another meaning was found, another action would be taken).

Higher the complexity of the system, higher the weight of memories/experience in the decision how and what to search, therefore – what could be found. It's a paradox, but:

More complex the system – More Meaningless the Meaning

The Smarter the system – the Meaningless the meaning

Because the system - the Artificial Intelligence or human - can more freely search and find meaning – links between the items, things, phenomenon, events, messages, memories, objects, images, sounds or whatever.

Continues with long comments after the article, about reinforcement learning and other topics... (to be published, when translated)

http://research.twenkid.com 

http://artificial-mind.blogspot.com 

http://eim.hit.bg/razum (Bulgarian)

...

This is my favorite picture from the time... :-P

Tosh in 2004 posing as "The Terminator"

Read More

Causes and reasons for human actions. Searching for causes. Whether higher or lower levels control. Control Units. Reinforcement learning.

Analysis of the meaning of a sentence, based on the knowledge base of an operational thinking machine. Reflections about the meaning and artificial intelligence.

Part 2 of 4 - Comment #1 and a part of #2 of:

Part 1 (и български): http://artificial-mind.blogspot.com/2010/01/semantic-analysis-of-sentence.html

One of the milestones of my AGI research. I wrote this particular article and the comments in Bulgarian as a 19-year old freshman in Computer Science at Plovdiv University.

By Todor Arnaudov | 13 March 2004 @ 21:49 EET | 340 reads |
First published at bgit.net and the e-zine “Sacred Computer” in Bulgarian

Comment #1 by Konstantin Spirov | 15 March 2004 @ 20:25 | EET

(…) I'm a classical programmer and I haven't really dealt with AI. However I reflected about how I would define “meaning”.

To me, the thirst and urge for finding a meaning does not prune contradictions, in contrary – it's searching for the cause, for the prime mover, the initial force. This is not related to contradictions.

For example, a tawdry clock with a thermometer is not incompatible, contradictory – it can hang on the wall, and neither the clock, nor the thermometer disturbs the other. However, to me this object is pointless, meaningless – you can't tell me any reason to put it on the wall.

On the other hand, the opposite phenomenon happens every day. People do absurd things, which however appear to be full of deep meaning. Programmers sing - out of tune... A banker who owns millions and visits luxury restaurants once passes next to an old lady who is selling donuts; then he takes one from a dirty bag and buys a donut for 30 cents. No doubt this donut was made by a poor snotty baker, but... For the banker, this is the best food and the best thing in the world!

There we are! For him, this is an act that is filled with deep meaning, but how would you persuade a computer program? Especially if the program is counting the number of viruses and bacterias that has entered the banker's organism in that very moment. How does the banker would explain it - “I felt a thrill, I remembered once when I was a child...”. From the viewpoint of the computer, this is a non-sense, just a random association – especially if from this little moment eventually grow up a serious decision for his life and career; how would you explain to a computer what's in common between the donuts and the money?

As we know, AI has many directions – the most of the researchers belong to the “weak” one, that is – not trying to model an AI, but just aiming to make the behavior of the computers to appear human, in order to make the life of the users easier – nothing more. These researchers discuss like this – the problem is complex and often vague, but we know tricks that would help us to cope it very well – we just have to spend some time and be more …. There are also researchers from the Strong direction, like the respectable Marvin Minskyor the clown prof. Kevin Warwick (sorry if you like him), who are aiming at goals which are much more interesting for the media. Some of them probably do, because of problems with funding, others do really believe, because this sounds more heroically.

I personally support the Weak AI and I think that the questions posed in this article very precisely describe the reasons – you can model non-contradictory system, learning and even something that looks like freedom (at least external unpredictability), but I cannot imagine how the thirst for a meaning can ever be modeled.

Scientists can invent any formal systems, to analyze and replay words, but we cannot give them a meaning (“да ги оглосим“ - an ancient Bulgarian word used). The meaning is a deeper concept than us, it is discussed by the ancient Greek philosophers. The whole human civilization from all the times deals with the meaning. (…) In the bible “слово“ (logos) means also meaning, cause. … “Logos“ - the First Cause, which can't be understood or explained (…other stuff about another explanation of a cat and an uphill about a real cat that has gone wild; usage of “to drink” in a metaphorical sense (to drink a stone with a gaze), hard to translate an not relevant:

P.S. за "котката и нанагорнището" ми хареса. Замислих се, че мога да дам неизчанчено обяснение за него. А компютърът - не.

Та сетих се, че за мен "котката изпи камъка и литна под нанагорнището" си е съвсем смислено изречение. Всеки, който си има котарак знае, че те са загадъчни същества и имат странни способности. Освен това мога да докажа, че съм виждал веднъж как котката пие камъка и лети под нанагорнището. Беше миналото лято на Варна. Тъст ми и тъщата ми си имат виличка на 30 километра от града, в една сушава местност с изглед към Варненското езеро. Пълен пущинак. Вилата име с картонени стени и се състои от две стаи.

Участникът в действието не е "котката", а една конкретна котка - котаракът Марти. Всяка дама би казала, че е и мил пухльо, http://polly-and-kosio.net/_predi/pages/bulkata_s_mama_i_tatko_jpg.htm, и в същото време, когато попадне в своята среда - той се превръща щастлив и див кръвопиец. До вилата има едно дере - когато Марти отиде на Варна, в него се събужда Хищника. Веднага избягва в дерето и по цяла седмица оттам не се чува нищо друго освен воя на вълците, лаенето на кучетата и крясъка на птиците. Точно, когато "родителите му" (тъст ми и тъщата ми) са изгубили всяка надежда, че ще го видят отново, той се завръща горд, с чувството на победител, ветеран, преживял своята неразбираема война.

Та изречението: "котката изпи камъка и литна под нанагорнището" много точно описва, какво се случи последния път, преди да изчезне в дерето. Зад камъка стоеше Иванчо, моят син, който току що беше проходил. Откакто "навлекът" бе там, Марти не получаваше достатъчно внимание. Преди да избяха на хълма и след това да литне под нанагорнището (към дерето), Марти за последен път изпи със завистлив поглед камъка, зад който се криеше Иванчо.

Regards, Kosio

#########

Comment # 2 by Todor Arnaudov | 18 March 2004 @ 21:58 EET | 0

Control Units, Causes, Goals, Achieving goals of a Control Unit == Pleasure

“Reasonable behaviour” - a search for local maxima of alterating and changing functions of expected pleasure and many more

[In brief, the concept of "Control unit" or CU means something like a causal force, it's more complex, but explanation is in the bigger theory of Universe and Mind to be translated.]

Thanks for the opinion and for the opportunity to post some more reflections on the topic... (...)

I agree that “meaning” has other meanings in different contexts and circumstances. E.g. “a purpose”, “goal”. “There is no purpose” means that I don't have a reason to do it, there is nothing that I want to achieve, linked to this “item”. With “reason” or “cause” - you cannot find a cause that could cause you to do this particular thing.

However, the first, primary causes are something with many features, as well. E.g.:

I feel thirst, I want to drink some water. Then, my goal becomes “to satisfy” my thirst. Then I start to search for means to achieve this goal in the possibly closest spatio-temporal area around me. I found this location is the sink, which is a few seconds away. I'm moving my chair a bit, get up, walk, open the door, pass through the corridor, open another door, turn around, take a cup, put the cup under the fountain, catch the tap for the cold water with my right hand, turn it; water spills; the cup gets filled, I turn the tap back; bend my hand back; prepare my mouth to drink; bend the cup, spill its content in my mouth; swallow....

Ready! The thirst was satisfied...

Machine: Why did he drink some water?

Human: Why, really I did?
Inpatient one: Because he was thirsty... It's obvious!

Machine: I don't think so. Why not the cause to be that because one or another Nuclear World power DID NOT send Hell just a moment before, so that he couldn't drink? Or why because there WAS NOT an earthquake etc. [Right, this is also because of Occam's Razor, complex examples – 2010 addition, but the point is that there are zillions of possible reasons and causes, and we're pruning them because we're searching for simple explanations.] There are simpler meaningful possibilities – if on the table next to him there was a bottle of a juice or another soft drink, he might haven't drunk water, but this beverage. In the situation's definition, it was not told that there was not such a bottle. This is an assumption, you don't even think that it is possible.

So it is possible that there was a bottle of soft drink, a Coke, and he has realized in this very moment, that those kind of soft drinks cause bad teeth and he has recalled the visit to his dentist. This is not mentioned in text, but it is not denied either.

And why not say that he drank water, because:
- He is a human? Or because...
- He is a living being? Or because
- It was hot? Or because...
- There was a schedule of the water supply, and in this very moment there was water in the tubes. Or.. because not him, but his throat was dry then!!!  //[dried then.] !!!! пресъхнало

Or because last day he did forget to fill the bottle that he keeps next to the computer.
And the reason, the cause not to fill the bottle was that previous night he was too much into commenting in an Internet forum. The reason to be so concentrated in this forum was that...

And so on... You, poor humans, there are endless number of possible reasons, not just those simple one, that you short-sighted humans see next to your nose.

Impatient one 2: Shut up, you stupid piece of metal. What the heck you know? You're a machine, machines can't think... Everything is formalized in you, so you're stupid. And there was a sentence... the more you know, the less you know.... Errr..

Impatiant one 3: He drank water, because he was thirsty. It's so simple!

Human: It seems so, because when one is searching for reasons, for causes, he is limiting the space of the search the way that fits his own desires in the particular case. If one has read “He wanted to drink some water”, the first and easy plausible explanation we see is written in direct words. Immediately, while one is creating the virtual world of this situation, he is setting precomputed, biased reasons, based on his initial impression. Then, when one seems to search for the causes, he finds them immediately – nice and easy, there are right there in the root of the search tree..

Impatient one 2: Did you believe this machine? You shouldn't! Never be persuaded by a computer, no matter as smart it seems to discuss. No matter how it appears to be, it cannot think, because it doesn't have a soul. I don't know what exactly a soul is, but .. Blah-blah...

Human: But you cannot drink water, if there is no a sink and a tap. The reason and cause he drank water was both his desire and the existence of a mean, a source to achieve its goal!

Machine: Good. And why did he bend your arm before he drank?

Impatient one: Why... In order to drink! You are so dumb!

Machine: But the arm doesn't know what is “to drink”. It is just a lever. He bended his arm, because his brain instructed the arm to bend. But this instruction was sent, because the human decided to drink, without bending the whole body and drink directly from the water jet. And there could be a reason not to do it, because he has a trauma in his back.... Why did he have the trauma? Because a few days ago he attempted to lift a too heavy weight; he did, because he wanted to practice, because he saw his girlfriend looking too much strong men's asses. Therefore, if we stop the search for a cause right here, then the reason, the cause that the man has bended his arm was his girlfriend, and her looks to the strong men's asses. However, why not saying that those strong men are the reason? The woman wouldn't look them, if they didn't exist. Or they muscles? The mere existence of muscles. Or these specific circumstances – in a specific moment they met strong men in the park, the woman looked at their asses, the man was jealous, he tried to lift heavy weights in order to practice, then this caused a trauma to his back, then he wanted to drink some water, and he used to bend his back and drink without a cup, but this time he couldn't, so that's why he bended his arm...

The conclusion is that there are many and many possible reasons and causes that are actually true in the same time, because the reason and cause for every event could be assumed to be everything that has happened, and depends on the moment we decide to stop and simplify. In the total, common cause there is no meaning – there is no specific reason/purpose/meaning/cause. The intelligent beings make sense and choose causes/reasons/purposes/meanings, based on their knowledge (and aim; knowledge = their biases/structure/configuration/state/development...).

(The intelligent beings are fitting reality to their virtual reality - fitting the laws in their virtual worlds to the laws they assume to be laws to the real world, comment from 2010).

Impatient one: The question was, what the brain has instructed the arm...

Patient one: Why what the brain has instructed? The cause the arm bended was that the muscles bended, then they pulled the bones, which are supporting the soft tissue of the arm and the hand, which is holding the cup.

Machine: Human call “reasons/causes/purposes” those ones that he himself, in particular, has accepted to call “causes”. In this specific case – the first items that appear in his mind. The first items that appear to a mind are his thoughts and feelings, linked to a particular event he recalls. And if a plausible enough cause/reason/purpose is found (usually “enough” is really a small ammount), the search is concluded and the searcher doesn't ask for more.

Patient: I see, but I'm tired already...

Konstantin: (…) There we are! For him, this is an act that is filled with deep meaning, but how would you persuade a computer program? Especially if the program is counting the number of viruses and bacterias that has entered the banker's organism in that very moment.

The persuasion depends on the both sides. If nobody can persuade you to do something means this only this, not that a reason/cause/purpose to do what he want you to do does not exist at all. If a man wants to do something, the most frequent reason he finds is... because he wants! Usually one doesn't know why he wants exactly this, or if he knows a little, he can “prove” it with explanations like “I want it, because I like to do it!”, “I enjoy it” or so. If, for some reason, one has to explain it in a more persuasive way, usually one searches for a plausible explanation why would someone want to do it and why would do it. If one doesn't want to do something, he says “I don't like to do something” - and always can find a reason why he don't want to do it.

If this imaginary program is intelligent, it could easily find many explanations.

Imaginary_AI: What a dummy program would deny that the act of the banker was meaningless? This banker, this is the richest man in Bulgaria and according to statistics, a few months later he has considerably enlarged his wealth. That means, he has reached a higher local maximum of his wealth (see below). Therefore, according to the behavioral model of his virtual control unit (see below) and the statistics, he has selected reasonable/meaningful actions and has taken good decisions in his spatio-eventual(based on events)-temporal region. His actions and decisions has led him to his the goal, which easily can be implied as “possessing more money”.

Imaginary_AI: There is no reason not to accept that buying the donut was not a part of his strategy to reach the general goal “being wealthier” (for example, it made him feel good, you give yourself such an explanation), because all and every actions and events, happening to a person, are linked and related to the way he thinks/reasons and to his following actions and decisions. All actions, done with a desire of the virtual control unit itself, intentionally and not forced - are meaningful and reasonable to itself. That means they are target actions, goals. Usually such target actions are caused by specific desires, initiated by a search of local maxima or high plateaus [of a reward]. My living friend will explain you this stuff below. [Reinforcement learning.]

Konstantin: To me, the thirst and urge for finding a meaning does not prune contradictions, in contrary – it's searching for the cause, for the prime mover, the initial force. This is not related to contradictions. For example, a tawdry clock with a thermometer is not incompatible, contradictory – it can hang on the wall, and neither the clock, nor the thermometer disturbs the other. However, to me this object is pointless, meaningless – you can't tell me any reason to put it on the wall.

It is possible that no one can tell you a reason to put in on the wall, because you think that it is tawdry and apparently this is "bad" and undesirable to you. Indeed - contradictions have to be searched and checked in the whole memory - of the evaluating unit (the agent, the human) together with the environment. [All history and possible relations.]

For example, one can find bad memories, related to such objects. [Which one does not realize, but they are fixed in the patterns of his mind.]

Konstantin: On the other hand, the opposite phenomenon happens every day. People do absurd things, which however appear to be full of deep meaning. Programmers sing - out of tune... A banker who owns millions and visits luxury restaurants once passes next to an old lady who is selling donuts; then he takes one from a dirty bag and buys a donut for 30 cents. No doubt this donut was made by a poor snotty baker, but... For the banker, this is the best food and the best thing in the world!

I don't think that any action or decision of any control unit is actually "absurd".

An evaluator calls it "absurd" when he or it doesn't really know the model of control unit's behaviour, or when the evaluator assumes that it knows how the evaluated unit/being "should behave" in a given situation.

However, if something unexpected or thought to be absurd/impossible has happened, it is apparent not that it is "absurd", but that the evaluator was WRONG. Either his model is wrong/not precise enough/confused or it could be precise, but it lacks the data required to make correct, complete and precise predictions.

If any Control Unit does anything, it has a particular meaning/reason/cause/goal behind it, even if it can be vague or unintelligible for an external evaluator.

The meaning/reason/cause is specific, it belongs to a particular working Virtual Control Unit. It is not a generalization, it is not a set of rules, written in a textbook. This is a specific model of something, that runs somewhere

The meaning of "meaning" right here is a GOAL. Any action of a Control Unit (CU), done because of instruction given by itself alone (and not forced by external CU, e.g. moving a hand with a wire) is tautologically target action for this Control Unit, and it displays the urge of the CU to achieve the "purpose/meaning/reason" of it to exist, according to its own understanding about what its goal is in the moment of decision. [Here "its own understanding includes also implied in the specific construction/architecture/the way the device/being works]

See my Teenage Theory of Mind and Universe for more. To be translated and published... http://research.twenkid.com

The purpose - the GOAL - especially for the compound CU is changeable and the more complete information about the exact event and circumstances we have, the more precise that GOAL could be guessed by an external evaluator.

A human are individuals in the sense that its indivisible - indivisible is what he understands "he is", but even theoretically he is incapable to know or understand exactly why it does what it does with the maximum possible resolution of control.

I think human [can be modeled as it...] is a complex of Control Units (virtual computers, simulators), where each of them is aiming at completing at maximum precision its program, the purpose of its existence (it's implied by its architecture and operation).

An indication of reaching to a goal of behaviour - finding of an optimum in the learning function - is the feeling of pleasure.

When a Control Unit detects that it has reached the goal, it “feels pleasure” and aims to fly about this part of the graph of its [reward] function.

Therefore I believe that human mind can be built as a mixture, a system of multilayer [hierarchical] control units, where each CU at a higher level controls with a lower resolution that the one below. The higher level control unit controls more imprecisely than the lower one.

For example, the top level of control sends a command with a length of 16 bits, while the description of the precise action to be done requires 128 bits or even... 2^128 bits. The details, the rest of those bits - 112 or 2^128 - 16 are actually completed by the "controlled" unit.
(It only seems that it's controlled, because the action is more dependent on its operation that on the operation of the top unit. [When the evaluations is done, evaluator probably usually starts from the top level, giving it “will”, initial cause... - comment from 2010]

(See "Abstract theory of the exceptions of the rules in computers", to be translated and published... http://research.twenkid.com )

A more specific example:

They say, that we can consciously control the moves of our fingers. Therefore our mind, our conscious can cause the finger to move...

Really?

We are free in the sense that when we think that if we want "I'll bend my finger right now!", then the miracle happens - exactly that finger bends. It seems that it moves, because of our free will. (This can be interpreted also as a coincidence, a match, and not a real control (causal relation), in terms of other articles from my theory from the time – comment from 2010).

This is power and control. However, in order the finger to bend in the reality and not just one to notice it in his mind, an enormous amount of information needs to be sent somewhere.

Not just selection of a finger (say 20 or 30 bits) and a definition of the a momentum, how strongly to bend the finger or so - this is just a virtual definition of a finger in our minds!

In the reality, the information that needs to be "entered" in order to execute that simple action includes the exact description of the precise movements and changes in every single particle that builds the finger, with the maximum possible resolution of the Universe.

Every single particle has a particular acceleration and it is in a particular place. Mind doesn't posses all that information and it can't, because it doesn't control in the strong sense of the world.

"Control" in its strong sense means with the highest possible resolution of control. [In the given environment/world/virtual world].

So measured as an amount of information, the cause of the movement of the finger is contained more in the finger itself than in the mind, the apparent control unit, because the description of the finger and the muscles that are acting on it is much longer than the description of the simple abstract instruction that we can realize and control [consciously].

Each CU assumes that it is the cause for the events to happen, that it is "free" and does what it wants, because each CU is similar to the only really free Control Unit - the whole Universe; it includes all details together. The whole Universe controls in the strong sense - what it "wants" happens, because it defines what is possible or not, and executes what's supposed to.

However, not every CU is complex enough in order to declare "I do control". Human mind is complex enough to do it, but actually the body is what controls it, not the reverse. The conscious can embrace only a part of the causes for its own existence, and no matter how hard it is searching for the deep ultimate causes of its own actions and decisions, it can't reach to them.

This is the free will, the freedom. The Control Unit (a human mind) cannot find the causes and reasons for its behaviour in the way [the precision, the domain ...] that it assumes that it should, if they had existed, and that's how the control unit proves that its own behaviour is free and unpredictable not only for an external evaluator, but universally.

Also, this is a convenient conclusion, when the CU intentionally aims not to find proves for the predictability, because one of the major goals of every CU is to feel as a MASTER. No matter how simple or complex (built by many simpler) the CU is. CU aims to feel as a MASTER and not to put this in doubt. CU are similar to Universe and they aim to be like it.

TO BE CONTINUED.... with part 3/4

http://research.twenkid.com
http://artificial-mind.blogspot.com
http://eim.hit.bg/razum (Bulgarian)

Other keywords: Universal AI, Twenkid Research

Read More

Motivation is dependent on local and specific stimuli, not general ones. Pleasure and displeasure as goal-state indicators. Reinforcement learning.

Analysis of the meaning of a sentence, based on the knowledge base of an operational thinking machine. Reflections about the meaning and artificial intelligence.

Part 3 of 4 - Comment #2 continues...

Part 1 (и български): http://artificial-mind.blogspot.com/2010/01/semantic-analysis-of-sentence.html

Part 2 (и български): http://artificial-mind.blogspot.com/2010/02/causes-and-reasons-for-any-particular.html

Part 4   Intelligence: search for the biggest cumulative reward for a given period ahead, based on given model of the rewards. Reinforcement learning.

One of the milestones of my AGI research. I wrote this particular article and the comments in Bulgarian as a 19-year old freshman in Computer Science at Plovdiv University.

By Todor Arnaudov | 13 March 2004 @ 21:49 EET | 340 reads |
First published at bgit.net and the e-zine “Sacred Computer” in Bulgarian

Comment # 2 by Todor Arnaudov | 18 March 2004 @ 21:58 EET |CONTINUES...

Let's note, that when a mind is being evaluated, a "stop frame" is evaluated. In this very moment. (...) And somebody with a particular attitude is the one who can find best why he does like or dislike a particular thing.

E.g. only you know, you see or imagine on what exact circumstances you're talking about. Every detail in a given situation is important, not only the generalized ones, said in a few sentences. E.g. on the wall that you imagine, there could be other particular objects which don't match this one. And this is your taste - your like or dislike.

The purpose of this "will", this kind of will - like/dislike something - is a search for a reason to decide and choose, when a single and unambiguous action should be executed. If that action doesn't lead to a damage (you won't be hurt either if that clock with a thermometer is put on the wall or not) - then there is not an immediate practical consequence/meaning what we choose. For a long period ahead, we cannot predict how exactly this particular action and decision will affect us, because the future inputs are too much and too much unknown in advance.

So, let's assume that it doesn't matter the clock & thermometer combo is on the wall or not, and the reason is that "we like it" (anyhow) and it is possible to find a persuasive reasonable explanation, if it is possible to analyze ourselves precisely enough.

It is possible that:

1. You already had a clock and a thermometer (separate) and you're a practical person who doesn't like possessing redundant stuff.
2. You don't need a thermometer anymore. - This is a preliminary calculated reason; once you have told to yourself "I don't need a thermometer anymore". Then you have never questioned this reason and you have followed it as a reason not to put a thermometer on the wall)
3. You're in a bad mood and you would deny anything that anyone would ask you to do.
4. You don't like anybody to tell you to do whatever, and you feel like they are giving you a command (e.g. it's a gift from your mother in law)
5. You just don't know why.


In general, of course - if we're searching for a reason, it's nice to have as complete model of the evaluator as possible...

Definition of a "meaning" with the meaning of a goal/purpose or cause/reason and "not contradictory link" expresses the attitude of the evaluator to the item that is being evaluated (where a cause/reason/purpose/meaning is searched).

And it is very important to note, that the evaluator determines whether there is or there is not a cause/reason/meaning/purpose for him personally. "Meaning" is subjective.

Actions of thinking machines and persons can't be explained unambiguously from an external evaluator, because externally the information is very scarce [and intelligent agents behaviour is very complex and non-linear]. The amount of information which is transferred between the parts/modules of the machine/human is enormously bigger than the data which is outputted - e.g. the externally visible behaviour.

Besides, the mind is biased and limits itself when searches. To me, the ultimate cause for anything is the whole past, the every single little difference would make the whole different. However, since it is impossible for the mind to compute all causes, mind uses greedy algorithms and searches for most direct and "plausible" explanations/causes/scenarios.

It is like differentiating in mathematics - there is raw data, a graph of values; the function that has drawn the function is searched. However, this is an ambiguous operation, we can guess, but not always know - the causes are also ambiguous.

That's why any "differentiated" causes are meaningful at a given deepness, resolution or so, where the search is interrupted.


Now, let's check Konstantin's opinion once more, differently:

Konstantin: People do absurd things, which however appear to be full of deep meaning. Programmers sing - out of tune... A banker who owns millions and visits luxury restaurants once passes next to an old lady who is selling donuts; then he takes one from a dirty bag and buys a donut for 30 cents. No doubt this donut was made by a poor snotty baker, but... For the banker, this is the best food and the best thing in the world!

Where's the absurd in programmer singing? "A programmer" means a person, and a person usually can sing - good or bad.
What you express here is just your opinion, your disapproval.

There is not a reason/purpose for a bad singer to sing? Why? Because he's afraid of being accused for singing, mocked... (...) 

[However] This particular programmer may be singing because:
- He feels in love and felt happier than the moment before, and singing is a way to express your good mood.
- He is alone, and he wanted to sing before, but he was shy.
- He is drunk and his inhibition was taken away.
- He is in a karaoke bar, lonely, he saw a beautiful girl and he decided that this is a meaningful way to attract her attention. He might be drunk or not.

(...)

It is impossible to embrace all possible reasons, because of the combinatorial explosion, but when having rich enough information about the circumstances, it is always possible to find plausible concrete reasons, if one wants to find one. (The one who denies, usually doesn't want to find reasons.)

Konstantin: For him, this is an act that is filled with deep meaning, but how would you persuade a computer program? Especially if the program is counting the number of viruses and bacteria that has entered the banker's organism in that very moment. How does the banker would explain it - “I felt a thrill, I remembered once when I was a child...”. From the viewpoint of the computer, this is a non-sense, just a random association – especially if from this little moment eventually grow up a serious decision for his life and career; how would you explain to a computer what's in common between the donuts and the money?

What does a "program" means? The implied in your words is "a dumb program" or one that would react as it doesn't understand. However, then obviously this is not a thinking machine!

I think the reason is actually very purposeful and meaningful. Every Control Unit has goals: the man has recalled an event that has made him feel good and he wanted to feel that pleasure again. He didn't think of the microbes, he didn't include them in the evaluation function, and he has never counted the microbes. If a machine is searching for a reason for a human, it must put itself in his shoes and evaluate as if it were him, not as if it was a counting machine. (As I noted before, at least according to my research, "pleasure" means achieving the goal of the behaviour of the Control Unit.

The banker has recalled a reachable state that in the past has made him feel pleasure.
That state is a set of circumstances, feelings, possibilities for actions; possible ways for changes in the circumstances/perceptions/feelings.

In that very moment there weren't any other possibilities that would give him higher pleasure in the next, say, 5 seconds. After he has bitten the donut, the search for another way to feel pleasure is cut. The donut became a "master" of the mind and rules the person's current goals and behaviour. [This local reward] rules the hands, the jaw etc. in order the person to feel the taste that has made him feel the pleasure in the past, and that pleasure to come back and be felt again.

How does this temporal control over the mind and the effectors (muscles) happen?

The behaviour of human can be represented as a complex of greedy algorithms which are searching for states, local maxima - the biggest possible pleasure, and the minimum possible displeasure.

Any chosen action is reasonable, for the virtual greedy algorithm that has ruled over the rest in the given situation. (...)

E.g. if one starts to eat some food and it's tasty, he doesn't spit it after the first bite, in order to search for something which is tastier, even if he knows that such food does exist and is near-by - in the fridge; the first bite may rule us for a moment and inhibits the urge for another piece of tasty food.

However, what if in the evaluation of the pleasure the mind includes "the fear of caries"? It is so complex mixture in the mind, what exactly would "rule" depends on specific memories, specific stimuli in the recent past and it may appear random.

In case of caries being included in the evaluation, it might be negative - the chocolate is not the maximum pleasure and shouldn't be taken. Different Control Units are working in parallel, all fighting for control over the effectors. And if one which has this fear rules out, it may stop the eating operation and switch to "brush my teeth immediately".

Before the caries consideration, the greedy algorithm has computed the biggest cumulative pleasure/reward in the near 1 second. However, the caries and the pain at the dentist forced him to look feelings long ahead, which are assumed to be caused by the teeth and chocolate. This assumption is important - something else may actually cause it, but the person takes this as a reason/cause!

The dental pain is much bigger punishment than not feeling neither pain, nor pleasure (not eating the chocolate), so it is avoided.

Turning back to the donut - this "random" link/memory/recall is not random at all!

The apparent reason that recalled the memory are the images of donuts, their smell. Also, the circumstance that the banker has been walking alone and he was thinking of something; not long ago he has met his grandparents in his village; when he was young he loved donuts...

All these details has made him want to feel this pleasure again right then.

Let's analyze the donut even once more:

Konstantin: People do absurd things, which however appear to be full of deep meaning. Programmers sing - out of tune... A banker who owns millions and visits luxury restaurants once passes next to an old lady who is selling donuts; then he takes one from a dirty bag and buys a donut for 30 cents. No doubt this donut was made by a poor snotty baker, but... For the banker, this is the best food and the best thing in the world!

The banker may possess millions, but while he's walking by the old lady on the streets, those millions are worthless. When one is walking on the street, he is supposed to follow the stimuli around him - reading the captions, watching the cars, traffic lights, passers-by. There are not luxury restaurants at every single corner, and you can't purchase a Ferrari or an airplane right there.

The action of the banker just seems "meaningless", i.e. inappropriate, impossible to explain, because it was assumed that if one agent is a banker, then he should do this and that, and never does this-and-that. This is another example of artificial self-pruning of the search of reason/purpose, without explaining why and without an explicit cause, besides the prejudice.

Every human being, even every humanoid robot can put his hand in a bag, catch the donut, pay and is capable to inform the others that "this is the great thing I've ever did in my life".

There is always a possible reason to do it, if this is the best/most rewarding action that the agent has found in the current situation/planning period.


CONTINUES with part 4/4...

More keywords: Universal AI, Artificial General Intelligence (AGI), Behaviorism, Psychology, Control Unit


Part 4   Intelligence: search for the biggest cumulative reward for a given period ahead, based on given model of the rewards. Reinforcement learning.

...



Corrections 3.Sept.2025, 3.9.2025:  "not as if it wa counting machine." (if it was a counting machine) - the style can be improved in some places as well.

* Note that there are some omissions from the complete text in Bulgarian (...), you can find the original from the first part of these English articles.

Visit the virtual conference Self-Improving General Intelligence SIGI-2025 and find the "Prophets of the Thinking Machines": https://github.com/Twenkid/SIGI-2025/

Read More

Ultimate AI, Free Energy Principle and Predictive Coding vs Todor and CogAlg - Discussion in Montreal.AI forum and Artificial Mind

Contents

1. The Interview - the key to true AI by the genius neuroscientist
2. CogAlg and Free Energy Principle
3. Discussion at Montreal.AI and the Ultimate AI
3.1. References to Bialek and Tishby early papers on prediction in RL
4. Ultimate Intelligence Part III ... - an informal review and a  clash of schools of thought
4.1. Intro and acknowledgments
4.2. Criticism
4.2.1. Sigma-Product-Log-Probability mathematical formula fetishism
4.2.2. Too general
4.2.3. Where's the hierarchy?
4.2.4. The sum of rewards and bounded rewards are obvious
4.2.5. The hierarchy as a deadlock breaker
4.2.6. Discussion on specific quotes
5. Conclusion

1. The interview - the key to true AI

A WIRED's interview with Karl Friston was getting popular recently in social media, claiming that the "genius neuroscientist might hold the key to true AI".

Initially it seemed interesting, maybe that was something new and revolutionary, since I've been quite ignorant not knowing of him - or maybe I have forgotten a long time ago?

Well, I took a look of how the topic of "free energy principle" and "predictive coding" is defined in generic sources such as Wikipedia.

The conclusion - yes, I agree, it's the right direction, another related school of thought, but I don't agree that these ideas are so grandiose or original as presented in the press*, they were quite obvious for "my school of thought" since it started around 2001-2004, when I was a 17-19 year old kid, a rebellious teenager who haven't read or cited the contemporary literature.

Edit: the proper recent technical/neuroscientific papers seem at a different level, though, better than the general directions and not that general and lacking hierarchy. Such as this one, suggested by Eray after he read this post. I haven't studied it yet and probably would comment later about this and other related materials:

Deep temporal models and active inference

Author links open overlay panelKarl J.FristonaRichardRoschaThomasParraCathyPriceaHowardBowmanbc

https://www.sciencedirect.com/science/article/pii/S0149763418302525?fbclid=IwAR1n1spCr7SFrSD9rEasY032FSBQnw9NbgfS-JbgctZ_SjKfOnjoqyWkwcg

---

* Sure, everything in the consumer-world-celebrity-driven media is exaggerated, glamorous, the genius, extraordinary, outstanding etc., that's not an exception.

Connecting general intelligence principles with physics/Universe trends and biology is not that unheard of as well. I assume maybe it was a surprise in circles of some kind of too specialized software developers or too practical RL-ists/maths/ML/developers who didn't care about philosophy, biology, cybernetics etc.


2. CogAlg and Free Energy Principle

I asked the owner of the CogAlg project Boris about his opinion, he said that he's been hearing about that theory "at least from a decade" and in short he didn't seem impressed, because it was "nothing novel".

As of myself, I think the explicit emphasis of the idea of reducing the space of states for the living organisms and intelligence is suggestive for people who face these ideas for the first time, however it's somewhat obvious for hierarchical systems and even simple "machines", as the gears, the pistons etc. serve as "sub-spaces" which limit and guide the space of possible states.

As defined in the most ancient basics of "my theory", the higher level patterns are constructed by selected sequences/sets of elements from the lower level which serve as "instructions" (discrete), therefore not all possible combinations are covered. Only a limited space is legal, which respectively reduces the search/combinatorial space of possibilities at the higher level, therefore it has "a reduced space". That's seen in the hierarchical structures in nature: atoms, molecules, cells, tissues etc.

That "free energy principle" is yet another proof that the general direction towards AGI are getting established in different domains by different researchers.

3. Discussion at Montreal.AI and the Ultimate AI

My criticism that this was not novel in a thread on Montreal.AI facebook page ended up in a discussion with Eray Ozkural - a researcher from Friston's school of thought, a fellow AGI researcher, an author of publications at the AGI conference and knowledgeable of the Reinforcement Learning literature - particularly more than me.

His term for AGI: "Ultimate AI".

It's adding one more to the list of:  A-General-I (AGI),  Universal-AI (UAI) , Strong-AI, General-AI, Human-level AI, Goedel Machine, ... "Versatile Limitless Explorer and Self-Improver" - VLESI (one of mine, if I remember correctly... :) ) etc.

See the original discussion by Todor and Eray:
A discussion about Free energy principle vs other theories about intelligence as prediction

He directed me to the III part of a series of his papers: https://arxiv.org/pdf/1709.03879.pdf

Ultimate Intelligence Part III: Measures of Intelligence, Perception and Intelligent Agents 

A nice title.

He mentioned also two pioneers of the prediction paradigm in RL of which I wasn't aware, prior the "early 2000s", the period I suggested: Bialek and Tishby

Papers with promising titles that pop up: Predictability, Complexity, and Learning
https://www.princeton.edu/~wbialek/our_papers/bnt_01a.pdf

The information bottleneck method: https://www.cs.huji.ac.il/labs/learning/Papers/allerton.pdf
https://arxiv.org/abs/physics/0004057

Submitted in 2000-2001, probably coming from late 90s.

...

4. Ultimate Intelligence Part III ... - an informal review


4.1. Intro and acknowledgments

I reviewed Eray's paper from my perspective and share comments of mine - as a clash of my "school of thought" with his/or theirs. Mine perhaps is more philosophical.

Overall, the paper is fine and I recommend it for studying if you like those "probability-log-maths" proves like in the papers of Hutter & Legg, Solomonoff's algorithmic probability and stuff. It also has good references, both to researchers and papers, which may give you a kickstart into the subject matter. That goes also for the list of other papers by this author, they have interesting titles, I checked only a few myself, though. Good work!

However I have general criticism to that "school", not personally to the author.

My first impression and general criticism is the mathematical formula fetishism, which is present in all kinds of papers like that. Maybe it's also LaTeX-one and towards those small fonts...

Summation, Product, Log, Probability, Wave functions?(the psy at the end) thus "phases" or/and just putting Greek and Latin letters for verbal/simple things, a - action, r - reward, ... Combination of them and...

There we are, everything seems solved or proved, and it passes as academic, goes to conferences.

IMO it's tautological in general. The sense denoted with these letters is defined with natural language words and it proves itself by its definition. It claims that "this is intelligence", computes something/minimize something etc. thus "it's solved".

IMO (to me) simple formulas, while required to represent the ideas "formally", are not much more insightful than defining the formula verbally, which usually is done anyway, above and below the formulas, since it's about such general matters.

On the other hand, it's not practical or is much more confusing to define verbally more specific or complex algorithms. They are not obvious as well and require a real computation with data to see where they go. In these cases it is required to write it in code

The math formulas from these "classical" kind of algorithmic probability papers do not grow up in complexity too much and are kind of obvious in their expected outcomes, because they are stuck at one line or a few lines and I can't see concepts growing up on that.

"Where calculation begins, comprehension ceases" - Schopenhauer.

I understand that this is probably desirable by their authors, but it's not quite incrementally insightful to me.

4.2. Too General

Goes for that school of Algorithmic Probability, Hutter's model etc.

AGI should be general, but not too general, because it turns into generalities or in the deep sea of practical or theoretical uncomputability.

I'm advocate of human-like seed-AGI which develops like a child and there are milestones that it's expected to achieve developmentally.

4.3. Where is the hierarchy?

I didn't find any mention of the word "hierarchy" or "levels" in the paper, while that's crucial in building a real and scaling generally intelligent system and RL agent, as explained below as well. It is also in the heart of many prediction-based or cybernetics schools, such as:

* Ray Kurzweil (I haven't read his "How to build a brain" book, but Eray mentioned the Hierarchical HMM as his approach)
* Jeff Hawkins (hierarchical tempory memory)
* Boris Kazachenko
* The deep learning community
* Preceded by earlier cyberneticians, notably Valentin Turchin and his book "The Phenomenon of Science".
* Edit+: Neuroscience itself, of course; the early Russian and Soviet research - Pavlov etc. Anohin  discusses feedback in 1935 (санкционирующая афферентацуя, later обратная афферентация) - prior to Wiener and the Cybernetics

Is the hierarchy implied  in the paper or other ones of the author as the process of search/adjustment of the highest sum of expected rewards etc.?

However how and when exactly the levels are spawned, separated and interfaced? How the "reward"is quantified for new levels, inter-levels? How the feedback is defined?
In fact that is one of the main questions of the real AGI, which would move it out of the "generalities" territory. Boris Kazachenko is trying to do it in his Cognitive Algorithm.


4.4. Sum of rewards, or The sum of rewards and bounded rewards are obvious

I think that the Sum of expected rewards for a selected period ahead as a measure of "intelligent" ("rational") behavior and the need for bounded reward are already not that special thing to say.

Yes, they have to be declared, but actually that was obvious back in the early 2000s. It seems it's known from the ancient times even from the economy and from human's greed and tendency towards more pleasure* and less displeasure.

In the academical part of the behavioral/psychological domain, the need to take into account that each single reward is or should be bounded for generally intelligent human agents is known empirically by Simon's satisficing and from the experiment with the rat that presses the lever to stimulate his "pleasure center".

It's known also in the everyday life by anyone, from observing the behavior in case of addiction, either in mild cases when one gets preoccupied with an activity, or in the severe cases of drug addictions.

The scientific part in the RL is that it writes explicit formulas and uses mathematical terms like "local minimum/maximum" or "endless cycle" - since when the reward of a particular action is too big, the agent is locked in an endless cycle or a local maximum/minimum.

However the phenomenon itself is obvious from the every-day experience. I am missing "grounding" and justifications out of the abstract formulas. Just formulas and optimization of some magnitude is tautology. I have similar criticism to CogAlg as well, even though it claims it has its justifications.

The need for a bounded reward is obvious even theoretically, because a local maximum reward or a "cycle" of actions with maximum local reward would and could limit catastrophically the range of input in which the agent searches, thus would limit the space of environments if it starts from scratch, therefore it would be "less general" and will get into too much "exploitation over exploration".

The bounded reward could be justified empirically both by the cases of addiction, as mentioned above, where an out of control magnitude of a "reward" (behavior drive) makes the victim a slave of too "narrow range" of repetitive goals; and also by the relations in a human society.  In general, locally, the extreme reward for one agent at the expense of the pain of many others is suppressed, except for the "elite", down to the masses.

Top-down relations and properties are different than bottom-up and at the same level, they are not symmetrical, but that's different topic.

Also no one can be "endlessly satisfied", there's a limit. One person, one "element", the mouth needs a little to stretch into a smile :), it couldn't stretch 10 times more.

 I know that the "maths guys" would laugh at these justifications, but presenting something obvious in simple formulas doesn't make it more meaningful, while inducing "formulas" from experience ("operator induction", or pattern discovery, predicting, modelling the input; conversion of representations between domains etc.) is what intelligence does.

The "money" or other resources could go with less of a limit or seemingly "endlessly", but they are abstract, the money are not mapped directly to the agent, rather are part of more complex systems in which the specific human agents are constituent parts. Such systems could be called "The Corporation", "The Capitalism", "The Economy" etc., but after a limit of "happiness" more money do not increase the general reward for the individual agents. For healthy and functioning human beings,  "happyness" is "computed" based on many more parameters, not just one, especially "the amount of money owned".

Indeed, IMO just the sum of (any) rewards is not "intelligent" (abstract, universal) per se, unless it's just self-defined like that, that this reward is intelligence, for some abstract reinforcement learning agent.

Prediction serves as a general definition and I agree with it, however it is also not enough if given alone, because like with the addiction, it could be cheated if defined too simply or if the agent goes into a space with locally-specific features allowing it to predict too easy.

That's why needs to include as a measure a *widening* of the range and the horizon of prediction and the formation of a generalization hierarchy.

It needs to be more complex.

4.5. The Hierarchy as a deadlock breaker

In order to avoid deadlocks of falling in a maximum/minimum hole, the hierarchical system should constantly project and act in varying time-slices and with varying reward models. A unified model would be an aggregation of those switching sub-models. See articles and slides from my works.

That implies that for a complex, hierarchical agent, *there is not one absolute best reward path*  and a measurement of intelligence, based just on the reward at the moment is right only in that window of comparison and the selected measures. It's not "objective", it's "best" for that specific selected model of the world and model of the rewards with specific limitations and compared to specific other trajectories, but in general and complex environments and multiple possible goals, there is a multitude of actions that have similar "rewards" or ones which keep the agent "alive" at a macro level. They are all "correct" and "intelligent". Thus intelligence needs to be more specifically defined with more parameters than just one "reward".

I don't like the definition of Hutter's quoted one: "the wide range of environments". If I'm not mistaken Ben Goertzel had something similar in the 2000s. IMO this is mundane, especially together with simple formulas.

Mapping it just to simple formulas of probabilities (varius kind ~ various pdf...) as a solution doesn't make it more clear. All kinds of papers like that look like Bayes or almost the same + - logP P(a,b) ... They resemble the basics of Shannon's Information Theory, which maybe has been one of my own inspirations for realizing that prediction and compression of information are the "keys to true intelligence".

A general flaw of that school is that these formulas are too general, indiscriminate, too universal, or as coined in this paper: "ultimate". It implies that they are also inefficient to calculate.

5. Notes on specific citations

"An adaptive system that tends to minimize average surprise (entropy) will tend to survive longer."

That said seems probably true, but only for a non-evolving system. Life as a whole "survives longer" by gradually adapting, trying new things and testing them for fitness, "evolving". At the moment of spawning new organisms when there's sexual reproduction, the exact combination of genes is unknown to the mother and father system, this is a big "surprise".

6. Conclusion

This article is underdeveloped, but that's it for now.

See also:

* The course program of the world first University course in AGI (see the links in the blog)
* Todor's Theory of Mind and Universe - his philosophy and principles, expressed in works from his teenage years
* Materials from the University course in Bulgarian and English: http://research.twenkid.com/agi/
* Анализ на смисъла на изречение и ...  March 2004, @ bgit
* Translated in English:

Analysis of the meaning of a sentence, based on the knowledge base of an operational thinking machine. Reflections about the meaning and artificial intelligence
http://artificial-mind.blogspot.com/2010/01/semantic-analysis-of-sentence.html

Part 1: Semantic analysis of a sentence. Reflections about the meaning of the meaning and the Artificial Intelligence

Part 2: Causes and reasons for human actions. Searching for causes. Whether higher or lower levels control. Control Units. Reinforcement learning.

Part 3: Motivation is dependent on local and specific stimuli, not general ones. Pleasure and displeasure as goal-state indicators. Reinforcement learning.

Part 4 : Intelligence: search for the biggest cumulative reward for a given period ahead, based on given model of the rewards. Reinforcement learning.

Many other articles from this Research blog - search them if you care, AGI digest, AGI email list, dscussion on the Cognitive Algorithm site etc.


* Note, 26.3.2023: A few corrections of found errors: Live - Life; "they are reminding ..." -- they resemble; "teenager years" - teenage years

Read More